SciELO - Scientific Electronic Library Online

vol.25 número1A reply to Julia Vergara's commentsA reply to Cheryl Chen's comments índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados



  • No hay articulos citadosCitado por SciELO

Links relacionados

  • No hay articulos similaresSimilares en SciELO


Análisis filosófico

versión On-line ISSN 1851-9636

Anal. filos. v.25 n.1 Ciudad Autónoma de Buenos Aires mayo 2005



Comments on Juan Rodríguez Larreta's paper: "conceiving of someone else's pain on the model of one's own"

Cheryl K. Chen,
Bryn Mawr College

Larreta's paper is concerned with what Norman Malcolm calls the "External Argument" concerning the problem of other minds.1 This argument is extrapolated from sections 302 and 350 of Wittgenstein's Investigations. The target of the External Argument, as I understand it, is the Argument from Analogy.2 According to the Argument from Analogy, I find correlations in my own case between certain mental states and certain kinds of bodily behavior. Then when I find similar kinds of behavior in other people, I have reason to conclude that their behavior is caused by similar kinds of mental states. For example, I have good reason to believe that other people feel pain because they behave the way I do when I feel pain. The core of the Argument from Analogy is the idea that I first learn what a sensation like pain is from my own case, and then I go on to ascribe pain to other people by supposing that they feel the same thing that I feel.
According to the External Argument, if I were to learn what pain is from my own case, then I could not intelligibly ascribe pain to other people. If this argument is correct, then either we are left with an extreme form of skepticism about other minds, or we do not learn what pain is from our own case. According to the External Argument, I cannot make the transition from my own pain to other people's pain because - in Wittgenstein's words - in order to ascribe pain to others I would "have to imagine pain which I do not feel on the model of pain which I do feel".3Wittgenstein himself says little about why felt pain is not an adequate basis for understanding unfelt pain. Malcolm thinks the point is that if I learn what pain is from my own case, I have learned that pain is something that is essentially felt by me. It then does not make sense to extend this concept of pain to the case of other people, since I do not feel other people's pain.4
Larreta thinks that Kripke gives a definitive objection to the External Argument. According to Kripke, there seems to be no reason why learning what pain is from my own case should entail that my concept of pain is of something that is essentially felt by me. In the same vein, if I learn what ducks are from encounters with ducks in Central Park, that does not mean that my concept of a duck is a concept of something that essentially resides in Central Park.5 Proponents of the External Argument need to say more, on Kripke's view, about why there is something illegitimate about extending the concept of pain from one's own case to the case of other people.
Because many philosophers still accept the External Argument despite Kripke's objection, Larreta offers an objection of his own. Larreta's objection is the following. The External Argument takes for granted that I can ascribe pain to myself, and this includes the ability to ascribe pains that I had in the past. Suppose we accept a popular contemporary theory of personal identity according to which my past self is not identical to my present self, but instead bears weaker relations of psychological continuity with my present self. So the difference between the relation between my present self and my past self and the relation between my present self and another person is just a matter of degree: the experiences of my past self have a greater degree of psychological continuity with my current experiences than the experiences of another person. Larreta concludes from this that if I can ascribe pains to my past self, then there is no reason to think I cannot also ascribe pains to other people.
I think Larreta raises some really interesting issues here, and I am especially intrigued by the connections he makes between the problem of other minds and issues about personal identity. The main point of my comments, however, is that Larreta's argument, while interesting, doesn't seem to add much to Kripke's initial objection. I worry that Larreta's argument doesn't go through unless we already accept Kripke's objection. But if we do accept Kripke's point, then Larreta's additional argument may be unnecessary.
Larreta makes an interesting point when he claims that ascribing pain to our past selves is no less problematic than ascribing pain to other people. I wonder, however, if this point could just as easily be taken as support for the External Argument - or at least a further nail in the coffin for the view that the External Argument is trying to refute. Remember that the target of the External Argument is the view that we learn what pain is from our own case, and then go on to apply the concept to other people when they behave in similar ways. The External Argument says that if we learn the concept of pain from our own case, then we cannot intelligibly ascribe pain to others. Larreta's point could be taken as a further problem for the view that we learn what pain is from our own case: if we learn what pain is from our own case, then we cannot legitimately ascribe pain to our past selves either. So much worse for starting from our own case. I think Larreta may be gesturing at this idea towards the end of his paper when he says that the basic problem of passing from the given to the non-given, or from the present to the absent, is not merely solipsism but, as he puts it, "solipsism of the present moment".
In the main part of his paper, however, Larreta grants what he takes to be the starting point of the External Argument: that we can ascribe pain to ourselves. He claims that such self-ascription should include ascriptions of pain to our past selves, and goes on to argue that we can extend the concept to others - since the difference between a past self and another person is just a matter of degree. Another person is someone whose thoughts and experiences are what he calls "zero-degree linked" to our own current mental life.
However, it seems question begging to grant that we can ascribe pain to our past selves. We cannot assume that we can understand unfelt pain on the basis of felt pain - that's presumably what Larreta wants to prove. But in that case, we cannot simply grant that we can ascribe pain to our past selves. After all, we do not feel the pain that we ascribe to our past selves any more than we feel the pain of other people. The External Argument asks how I can say that another person has the same pain that I have, when my pain hurts me and the other person's pain does not. In the same vein, how can I say that I had the same pain yesterday that I am feeling now, if the pain I am feeling now hurts while the pain I had yesterday does not?
I think this is a point we can make even if we do not buy into a Neo-Lockean theory of personal identity. The point may still hold even if we suppose that there is, in fact, a self or ego that endures through time. We can allow that I am the same person now as the person who felt pain yesterday; the point is that there is still something illegitimate about ascribing pain that I do not now feel- even if I am ascribing it to myself.
Of course, Larreta could at this stage remind us of Kripke's objection: the External Argument does not give us any reasons to suppose that we cannot understand unfelt pain on the basis of felt pain. However, if we are prepared to grant Kripke's objection, then Larreta's move seems like an extra wheel. It looks like Kripke's objection is what is really doing all the work here. Ifwe allow that it is permissible to extend our concept pain to include unfelt pain, as Kripke suggests, then the problem of ascri bing pain to other people does not even arise. There is no need to appeal to any parallels between other people and past selves.
Perhaps Larreta could insist that the External Argument's point about understanding unfelt pain on the basis of felt pain doesn't apply to the case of pains I have had in the past. One could try arguing here that I can ascribe past pains to myself as long as I remember feeling those pains. This may be right, but I do not think it will help Larreta's line of argument. For then it seems we would have to accept that there's an important difference between ascribing past pains to myself and ascribing pains to other people. In the former case, I remember the painful feeling; in the latter case, I don't. And now it seems much more problematic how we can move - as Larreta says we can - from ascribing pain to one's past self to ascribing pain to other people.
Here Larreta might remind us of his discussion about the possibility of fusions. Fusions are hypothetical cases where two people become one.6 After the fusion, the resulting subject can recollect "from the inside" events that belong to the lives of both of his ancestors - while recognizing that these events belong to different subjects. For example, if I am fused with George W. Bush, I will be able to recall events from Bush's life as if they were my own memories; and yet I still recognize that these memories are of events that happened to Bush, not me. Larreta argues that if it is possible to recall past experiences of other people "from the inside" - in particular, if it is possible to recall other people's past sensations of pain - then ascribing pain to other people shouldn't be that much more difficult than ascribing pain to our past selves.
The worry I have here is that even if we allow that we could ascribe pain to people whose experiences of pain we can recall "from the inside", that does not mean we could ascribe pain to people whose experiences we cannot recall in that way (and I take that to be the norm in the actual world). Larreta wants to say that we can make that transition because the difference between myself, my past self, the past self of a person I am fused with, and a completely distinct person is only a matter of degree. It is just a matter of higher or lower degrees of psychological continuity: the experiences of a completely distinct person have what Larreta calls a "zero-degree" relation of continuity with my own. But I take it that zerodegree continuity just means no continuity. It is not clear to me why it is legitimate to ascribe pain to people whose experiences of pain are completely inaccessible to me.
I by no means want to endorse the External Argument. What I do want to stress is that in evaluating the External Argument, we should focus on whether Kripke's objection is correct. Are there any good reasons for thinking that we cannot legitimately extend our concept of pain from felt pain to unfelt pain? Is the case of learning what pain is from one's own case really analogous to learning what ducks are from ducks observed in Central Park? The answer to these questions should affect the legitimacy of ascriptions of pain both to other people and to our past selves.


1 Malcolm, (1963, pp. 96-129).

2 See, e.g., Russell (1998, chapter 8).

3 Wittgenstein (1953, section 302).

4 Malcolm (1963, pp. 105-106).

5 Kripke (1982, pp. 116-117).

6 Parfit (1975, pp. 199-223).


1. Kripke, Saul (1982), Wittgenstein on Rules and Private Language, Cambridge, Mass.: Harvard Universtiy Press.        [ Links ]

2. Malcolm, Norman (1963), Knowledge and Certainty, London: Cornell University Press.        [ Links ]

3. Parfit, Derek (1975), "Personal Identity", in Personal Identity, edited by John Perry. Berkeley: University of California Press.        [ Links ]

4. Russell, Bertrand (1998), Problems of Philosophy, Oxford: Oxford University Press.        [ Links ]

5. Wittgenstein, Ludwig(1953), Philosophical Investigations, Translated by G.E.M. Anscombe. Oxford: Blackwell.         [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License